NVIDIA RTX Spark AI PCs and Workstations at Scan

NVIDIA RTX Spark™

A new beginning

Pre-Order Now

The future of AI PCs Starts Here

NVIDIA RTX Spark™ introduces a new generation of AI-powered computing, bringing together NVIDIA AI, RTX graphics and Windows in a revolutionary new platform designed for creators, developers, professionals and gamers.

Built around the NVIDIA RTX Spark Superchip, these next-generation systems combine exceptional AI performance, advanced RTX graphics and incredible efficiency in ultra-thin laptops and compact desktop PCs.

Shop By Brand

Shop Now

Ignite Your Spark

Experience RTX Spark in the Cloud and explore it's capabilities in a flexible hands-on environment.

Try RTX Spark in the Cloud free for 48 hours before you purchase and experience the power of the new RTX Spark Superchip on your projects

Powered by SCAN Cloud

Shop Cloud

Your Passion. Your Spark

Explore how NVIDIA RTX Spark can accelerate gaming, creative and professional AI workflows across Scan's specialist Business Units.

Spark for Pro Gaming

Experience higher frame rates, richer visuals and more responsive gameplay with NVIDIA RTX Spark. DLSS and Multi Frame Generation boost performance with AI, while advanced ray tracing delivers realistic lighting and detail. NVIDIA Reflex keeps response times low, and integrated technologies help balance powerful gaming performance with battery life when you’re away from the desk.

Spark for Pro Audio

Handle larger sessions with more instruments, plugins and effects, while keeping your DAW responsive and fluid. AI-powered audio tools can run locally for faster noise removal, stem separation and restoration, while powerful multitasking lets you work across your DAW, sample libraries and other creative apps with ease. All that performance comes in a portable package, giving you the freedom to take your studio setup wherever the session takes you.

Spark for Pro Video

Edit, preview and export high-resolution video with greater speed and responsiveness. Dedicated NVIDIA encoders accelerate exports, while GPU acceleration keeps demanding timelines smooth and responsive. AI-powered tools for masking, tracking, enhancement and noise reduction can run faster, helping you spend less time waiting and more time creating. From multiple high-resolution streams to final export, powerful hardware helps keep your workflow moving with fewer processing bottlenecks.

Spark for AI

Develop, test and run AI models locally without always relying on cloud infrastructure. GPU acceleration helps speed up model development, computer vision and robotics simulation, while high-memory architecture makes it easier to work with demanding models and datasets. From writing code and testing models to simulating physical AI, take your development environment wherever your work takes you.

RTX Spark Superchip

NVIDIA RTX Spark Superchip

Up to

6,144 Core

Blackwell RTX GPU

Up to

20 Core

Ultra-Efficient CPU

Up to

1 Petaflop

FP4 AI Performance

Up to

128 GB

Unified Memory

Built for Agents and AI

CUDA, the software that accelerated the world's AI, runs natively on RTX Spark.

All-Day Battery Life

The most power-efficient RTX chip ever made, in a chassis so slim you'll forget you're carrying it.

Creator Crafted

Make a million ideas happen with hundreds of creative apps and AI tools supercharged by RTX and NVIDIA Studio tools.

Game Ready

Play the latest and greatest games with world-leading gaming tech. Ray-traced worlds, the full DLSS suite, NVIDIA Reflex and G-SYNC.

What Does More Memory Unlock?

Traditional Windows laptops split memory into two fixed pools: system RAM for Windows and applications, and VRAM for the GPU. RTX Spark uses unified memory instead, creating one shared pool that can flex around gaming, creativity and local AI.

Built for how you use a laptop today, and how AI will change it tomorrow.

8GB

Good Performance, Limited Headroom

8GB is common in mainstream gaming laptops and is enough for many games, everyday creative apps and smaller local AI models. But newer AAA games and larger AI models can quickly exceed 8GB.

Best for: Mainstream gaming and light local AI

Limitation: Memory becomes the bottleneck before GPU speed does

Memory Available8GB of 128GB Maximum

Great for Mainstream Gaming

Everyday Creative Apps

Smaller Local AI Models

Unified memory is shared, so Windows and running applications still use part of the available capacity. The advantage is flexibility rather than having the full 8GB permanently dedicated to the GPU.

16GB

More Room for Gaming and Local AI

16GB gives demanding games and creative applications much more breathing room, while opening up a wider range of useful local AI models for writing, coding and image generation.

Best for: Enthusiast gaming, creation and practical local AI

Step Up: More headroom in today's games and more capable AI on-device

Memory Available16GB of 128GB Maximum

Gaming at Higher Settings

Content Creation

More Capable Local AI

Unified memory is shared, so Windows and running applications still use part of the available capacity. The advantage is flexibility rather than having the full 16GB permanently dedicated to the GPU.

24GB

Unified Memory Starts Here

A traditional laptop with 16GB RAM and an 8GB GPU may also have 24GB in total, but the GPU is still capped at 8GB. RTX Spark shares one 24GB pool across Windows, apps, the CPU and GPU, giving much more flexibility for gaming and local AI.

Best for: Gamers moving on from 8GB or 12GB GPU limits

Key Benefit: More flexibility for both modern games and local AI

Memory Available24GB of 128GB Maximum

Traditional Laptop16GB system RAM for Windows and apps, and 8GB VRAM fixed to the GPU.

24GB Total Installed, but the GPU Is Limited to 8GB.

RTX Spark24GB unified memory in one pool for Windows, apps, CPU and GPU.

Unified memory is shared, so Windows and running applications still use part of the available capacity. The advantage is flexibility rather than having the full 24GB permanently dedicated to the GPU.

32GB

Flagship Consumer Memory Capacity

32GB matches the memory capacity of an NVIDIA RTX 5090 desktop GPU. In an RTX Spark laptop, this level of unified memory gives you substantially more room for larger AI models, longer context and demanding creative workloads.

Best for: High-end PC capability with serious local AI

Step Up: More space for larger models and more breathing room for AI and apps together

Memory Available32GB of 128GB Maximum

High-End Gaming

Serious Creative Workflows

Run Larger Local AI Models

Unified memory is shared, so Windows and running applications still use part of the available capacity. The advantage is flexibility rather than having the full 32GB permanently dedicated to the GPU.

48GB

Beyond Conventional Consumer GPUs

48GB moves beyond the memory capacity of any current consumer GeForce GPU. It gives you much more room for larger local AI models and heavier workloads while still leaving room for gaming, creation and everyday apps.

Best for: One premium laptop for gaming, work, creation and advanced local AI

The Choice: Maximum GPU performance or much more AI memory headroom

Memory Available48GB of 128GB Maximum

RTX 5090 LaptopMaximum GPU performance and 24GB dedicated GPU memory. The fastest choice for traditional GPU workloads, such as gaming and content creation.

Maximum GPU Performance or Much More AI Memory Headroom

RTX Spark 48GBHigh GPU performance and 48GB unified memory. More AI headroom for significantly larger models.

Unified memory is shared, so Windows and running applications still use part of the available capacity. The advantage is flexibility rather than having the full 48GB permanently dedicated to the GPU.

64GB

AI alongside Everything Else

At 64GB, the question starts changing from 'Can this model fit?' to 'How much AI can I run alongside everything else?' There is enough headroom for advanced local models, agents and supporting AI services while you still game, create and work.

Best for: Users who expect AI to become part of their everyday PC experience

Unlocks: Advanced assistants, local agents and multiple AI workloads at once

Memory Available64GB of 128GB Maximum

Game and Create at the Same Time

Run Advanced Local AI and Agents

Keep More Apps and Workloads Open Together

Unified memory is shared, so Windows and running applications still use part of the available capacity. The advantage is flexibility rather than having the full 64GB permanently dedicated to the GPU.

96GB

Workstation Reference

96GB is not an RTX Spark configuration. It is included as a reference point because this level of GPU memory is normally found in professional workstation hardware such as the RTX PRO 6000.

Best for: A reference for where laptop memory begins to overlap with workstation-class capability

Why It's Here: Shows how far RTX Spark pushes beyond traditional laptop memory

Workstation GPU Memory96GB of 128GB Maximum

Very Large Models and Datasets

Professional AI Workloads

Advanced Rendering and Simulation

96GB is shown as a reference point and is not an RTX Spark configuration. It represents the level of GPU memory typically found in professional workstation hardware.

128GB

Maximum Local AI Headroom

128GB fundamentally changes the ceiling of what can live locally on a laptop. Very large AI models, sophisticated agents and multiple AI systems can coexist alongside Windows, games and creator applications.

Best for: Users who want the strongest possible local AI capability and maximum room for what comes next

Big Idea: Run more intelligence at the same time

Memory Available128GB of 128GB Maximum

Largest Local AI Models

Multiple Agents at the Same Time

Ready for What's Next

Unified memory is shared, so Windows and running applications still use part of the available capacity. The advantage is flexibility rather than having the full 128GB permanently dedicated to the GPU.

Shop RTX Spark Laptops
AI Agents

AI Agents

Your AI Assistant, Built into Your PC

AI agents are changing how we work. Instead of simply responding to commands, they can plan tasks, automate workflows, generate content, write code and help solve complex problems, all from your desktop.

With NVIDIA RTX Spark, you can run powerful AI agents locally, giving you faster responses, greater privacy and more control over your data.

Whether you're developing software, creating content, analysing data or streamlining business processes, AI agents help you spend less time on repetitive tasks and more time on high-value work.

Find Out More

Make More, Code More, Play More

Feature card showcasing product capabilities

Creators

Your Creative Power, Unlocked

Purpose-built acceleration for AI, 3D design, video production and live streaming, delivering faster workflows and professional-quality results.

Service highlights and benefits overview

Developers

The AI Standard

Built on the same NVIDIA CUDA platform that powers today's leading AI models, RTX Spark lets you develop, test and run AI workloads locally with up to 128GB of unified memory.

Case study presentation and results

Gamers

Game Ready

Experience the latest games with NVIDIA RTX technologies, including ray tracing, DLSS and NVIDIA Reflex for stunning visuals, higher frame rates and ultra-responsive gameplay.

Why Scan?

As one of the UK's leading NVIDIA Elite Partners, Scan combines access to the latest NVIDIA technologies with decades of expertise in high-performance computing, AI and professional workstations.

When you choose Scan, you benefit from:

  • NVIDIA-certified expertise
  • Dedicated AI and workstation specialists
  • Pre-sales guidance and technical consultation
  • UK-based support and customer service
  • Access to the latest NVIDIA innovations
  • Trusted by businesses, creators and professionals across the UK

From AI-powered PCs to enterprise-scale infrastructure, Scan helps you get the most from your NVIDIA investment.